Hello,

I have an unRAID NAS that hasn't burped in 4 years...til today. I'm not super experienced at linux, and, per the unRAID documentation ("Single parity risks: If you have single parity, you're at risk of data loss if another drive fails during the upgrade. Seek advice in forums if this happens.") Now that I see the Disk 1 in the attached, I am seeking advice.

The snapshot does NOT reflect that Disk 1 had 1024 read errors. I panicked when I saw the red X and powered down the NAS so I could do some reading, so the error count got zeroed. I then ran a short SMART self-test and got the following:

"Completed without error"

I am very willing to replace the drive that is shut down, but because of the above warning about single parity, I don't want to screw up the replacement. What should be my next step? Should I run an extended SMART self-test? Should I run a parity check? I have never had any errors in a parity check. The last one ran fine:

Last check completed on Sat 31 Jan 2026 01:38 AM (twenty-five days ago)

Duration: 17 hours, 29 minutes, 44 seconds. Average speed: 127.0 MB/s

Finding 0 errors

Thank you very much in advance from a noobie.

...Steve

Attach Diagnostics ZIP to your NEXT post in this thread.

Thank you for such a prompt reply. The anonymized diagnostics are attached below.

smg-nas-diagnostics-20260225-1053.zip

@trurl

Well, I just read the "Need help? Read me first!" link in your signature line.

Lesson 1: Almost always, the most important thing you need to do is capture your complete syslog, BEFORE YOU REBOOT! We usually need to see what went wrong BEFORE the reboot, because once you reboot, it's lost!

So, I've screwed up already....sigh.

SMART report attached

smg-nas-smart-20260225-1105.zip

I should probably update my sig links to ask for diagnostics instead.

Unrelated, your system share has files on the array. Possibly these were created because you enabled Docker/VM Manager without cache. We can look at that later.

SMART for disk1 looks OK. Emulated disk1 is mounted.

Looks like it was a connection problem, and disk2 looks like it is having the same.

All disks are on the same controller. How are these disks powered? Any splitters?

Oh gosh. It's been 4 years. I will power it down and look for power splitters. I'm pretty confident there aren't, but I will check. Please stand by.

Check all disk connections, data and power, both ends.

No splitters. Connections are good.

On Disk 1, the dead device, the dashboard says

Device is disabled, Contents emulated

Click to spin down device

I don't know if that's helpful.

Disk1 is not dead, just disabled.

2 hours ago, trurl said:

SMART for disk1 looks OK. Emulated disk1 is mounted.

Disk1 has

# 1  Extended offline    Aborted by host               90%     27331         -

Did you abort the extended self-test? 8TB will take many hours, and it only updates every 10%

Yes. I thought I'd better do as you requested, so I stopped it and checked the connections. Should I restart it?

1 minute ago, StevieGee said:

Should I restart it?

Yes, might as well do disk2 at the same time. It should finish a lot sooner since it is much smaller.

OK. They are both started. Is there anything I should do after that or just send the two reports?

Might as well do them all at the same time, some are getting old.

got it. they're all running now. I'm thinking I should replace the drives once the data are built back/safe. Do you agree?

Also, should extended tests be run periodically?

Edited February 25Feb 25 by StevieGee

If any don't pass you must replace, but you can only replace one at a time. Let's wait and see.

Results attached for all. Here's what I think, but I'd appreciate you proofreading:

All the drives have prefailure flags. Are they all needing to be replaced?

Parity is good

Disk1 doesn't say failure, but it seems to be in communicado. I have two 8TB HDDs that are supposed to be here tomorrow.

Disk2 restarted. The host interrupted the test. How the host interrupts one disk but not all, I don't understand.

Disk3 OK

Disk4 OK

Again, thank you so much for your help.

Disk3-smg-nas-smart-20260226-0903.zipDisk1-smg-nas-smart-20260226-0902.zipParity-smg-nas-smart-20260226-0900.zipDisk2-smg-nas-smart-20260226-0904.zipDisk4-smg-nas-smart-20260226-0904.zip

Not clear any need replacing.

Disk1 does need to be rebuilt. No real reason to think it can't be rebuilt to the same disk though, which is already 8TB so replacing it with 8TB doesn't gain any capacity.

Disk4 is the oldest and only 2TB, so replacing it would give additional 6TB.

Disk3 is the newest.

Disk2 is only 2TB so replacing it would also give additional 6TB.

Do you have another copy of anything important and irreplaceable? Parity is not a substitute for backup.

@trurl

Again, I can't thank you enough for your help.

Attached is the Disk2 extended smart report, which unRAID says completed without error. (I need to do some reading about this tool. The text of the report has a lot of stuff that looks like errors, so I need to learn a little more about smarttool interpretation).

Also, yes, I do have a backup of my data files to iDrive via Duplicati. I'm pretty sure that the backup is good. I have not tested a complete restore, but random file restores are successful. I also have a somewhat recent copy to a USB HDD.

Here's what I think I need to do now.  I'm planning this from unRAID doc located here

1. The doc says I should run a parity check.  Should I do that?

2. Rebuild the disabled Disk 1 in place per the doc.

   1. The doc talks about doing an "automated xfs repair". The criteria for doing that are not clear to me. I've used file explorer from my pc to look at some files and they appear good.

3. Run a parity check.

4. Stop the array.  Remove an old empty data drive from the array. Power down, replace with new drive. Power up. Assign the new drive as a parity device.

5. Run a parity check.

Does that sound right?

Also, you said, "Unrelated, your system share has files on the array. Possibly these were created because you enabled Docker/VM Manager without cache. We can look at that later."

How do I tackle that?

Disk2-smg-nas-smart-20260227-0910.zip

Those are (sort of) the correct docs you linked, but the steps you outline seem to have mixed some things up. And we are really going to be working from much later down in the docs.

Since you have single parity and a disabled disk, it can't actually do a parity check. I think if you go to Array Operation, it will just let you do a Read Check, which only checks that all remaining disks are readable.

According to those diagnostics, emulated disk1 is mountable, and as you say, you can examine the files. So shouldn't need repair.

Step 4 is way off. We aren't rebuilding parity, and we aren't removing anything. We are going to rebuild disk1 onto the same disk.

https://docs.unraid.net/unraid-os/using-unraid-to/manage-storage/array/replacing-disks-in-array/#re-enabling-a-disabled-disk-rebuilding-onto-itself

Got it.

The rebuild finished and the array is up. I'm freshening backups at this point.

My plan is to add a second parity drive when it arrives.

As you said, "Unrelated, your system share has files on the array." I'm thinking you're seeing the /isos and /system directories on disk2. Do I need to correct that? I am not sure how those directories got there.

Post new diagnostics

You got it.

smg-nas-diagnostics-20260228-1236.zip

Ideally, Docker/VM related shares - appdata, domains, system - would have all files on cache or other pool with nothing on the array, so Docker/VM will perform better, and so array disks can spin down since these files are always open.

Your system share has files on disk1 and disk2 and cache. Not sure why it would be on 3 drives since normally that share only has 2 files.

It is configured to be moved to cache, but nothing can move open files. You have to go to Settings and disable both Docker and VM Manager before that share can be worked on. And mover won't overwrite files. If there are duplicates you will have to decide which to keep.

Do you know how to use the built-in File Manager to examine your shares, disks, and pools?